OleehyO
0cba17d9ce
[refactor] Init
2025-04-19 14:32:28 +00:00
OleehyO
e0cbf2c99f
[chore] Cleanup
2025-04-17 07:08:47 +00:00
OleehyO
4e2740ada0
[feat] Support n-gram stop criteria
2025-04-02 03:23:27 +00:00
OleehyO
509cb75dfa
[deps] Change onnx-gpu to manually install
2025-04-02 02:48:23 +00:00
三洋三洋
5673adecff
[feat][formatter] Integrate LaTeX formatter for improved formula readability
...
- Add latex_formatter.py based on tex-fmt (https://github.com/WGUNDERWOOD/tex-fmt )
- Update to_katex.py to use the new formatter
- Enhance LaTeX formula output with better formatting and readability
This integration helps make generated LaTeX formulas more readable and
maintainable by applying consistent formatting rules.
2025-03-01 00:55:41 +08:00
三洋三洋
5cda58e8fc
[chore] Ignore ruff lint E741
2025-03-01 00:54:57 +08:00
三洋三洋
c35b0e9f53
[fix] Add project prefix
2025-02-28 23:38:12 +08:00
三洋三洋
f023c6741e
[feat] Remove bold style
2025-02-28 23:38:12 +08:00
三洋三洋
be9e32b439
[deps] Add ray serve & python-multipart
2025-02-28 23:37:53 +08:00
三洋三洋
cd9e4146e0
[chore] Add build system and pakage location
2025-02-28 23:18:06 +08:00
三洋三洋
3e9c6c00b8
[chore] Add python related rules
2025-02-28 23:18:03 +08:00
三洋三洋
0e5c5fd706
[chore] Remove unsed files
2025-02-28 20:54:51 +08:00
三洋三洋
4d3714bb4b
[chore] exclude paddleocr directory from pre-commit hooks
2025-02-28 20:01:54 +08:00
三洋三洋
3296077461
[chore] Setup project infrastructure
2025-02-28 20:01:52 +08:00
三洋三洋
a0942db712
[deps] pin transformers to 4.45.2 and sentence-transformers to 3.1.1
2025-02-01 13:00:44 +08:00
OleehyO
cee83611b5
Merge pull request #78 from OleehyO/pre_release
...
Change to better import dependency
2024-08-07 12:43:15 +08:00
三洋三洋
e1046ba3fa
Change to better import dependency
2024-08-07 01:19:26 +08:00
OleehyO
bbc8ecf88b
Merge pull request #67 from OleehyO/pre_release
...
Change setting name
2024-07-11 20:34:50 +08:00
三洋三洋
7438dee7ac
Change setting name
2024-07-11 20:33:51 +08:00
OleehyO
be922cc952
Merge pull request #60 from OleehyO/pre_release
...
Pre release
2024-06-23 22:16:09 +08:00
三洋三洋
bfb1810fb0
Update README
2024-06-23 22:14:05 +08:00
三洋三洋
838febf48c
Remove onnxruntime-gpu
2024-06-23 22:13:51 +08:00
OleehyO
69f53d7256
Merge pull request #59 from OleehyO/pre_release
...
Pre release
2024-06-22 23:56:45 +08:00
三洋三洋
6793142557
Update model config
2024-06-22 22:08:08 +08:00
三洋三洋
25f6cddf72
Update README
2024-06-22 22:00:14 +08:00
三洋三洋
cd519d8e99
Support onnx runtime
2024-06-22 22:00:05 +08:00
三洋三洋
2ae59776fa
Add optimum
2024-06-22 21:49:47 +08:00
OleehyO
529fba4db6
Merge pull request #58 from OleehyO/pre_release
...
Add formula detection service
2024-06-17 21:26:35 +08:00
三洋三洋
d8659cd3a9
Add formula detection service
2024-06-17 21:23:55 +08:00
OleehyO
18dc6497ae
Merge pull request #56 from OleehyO/pre_release
...
Add docker link
2024-06-11 13:22:17 +08:00
三洋三洋
c849728ee7
Add docker link
2024-06-11 13:20:32 +08:00
三洋三洋
a1c2b5b1ef
Update server.py
...
1. Change the default host address to 0.0.0.0.
2. Convert the output to KaTeX.
2024-06-07 12:26:24 +00:00
三洋三洋
6fbd285658
Update README
2024-06-07 06:54:23 +00:00
三洋三洋
9f4058c64b
Add Apache2.0 license
2024-06-06 13:06:16 +00:00
三洋三洋
236489ba2a
Add cover.png
2024-06-06 13:06:16 +00:00
三洋三洋
2920b753a8
Modify the names of options in the web.py
...
Formula only -> Formula recognition
Text formula mixed -> Paragraph recognition
Improved display during mixed inference
2024-06-06 13:06:16 +00:00
三洋三洋
dbbec511ef
Refine mix_inference
...
1. Add the formula number back to the isolated formula and merge multiple tag.
2. remove bold effect from inline formuals
3. change split environment into aligned
2024-06-06 13:06:11 +00:00
三洋三洋
29e626c984
Bugfix: to_katex.py
...
1. Added `change_all` function to fix a bug where some LaTeX formulas with the same wrapper were causing issues.
2. Removed some unnecessary formatting commands.
Bugfix: to_katex.py
2024-06-06 08:25:50 +00:00
三洋三洋
848726e6e2
Update
2024-05-28 09:51:53 +00:00
三洋三洋
e66f237cfd
Added releasing file
2024-05-28 07:50:09 +00:00
三洋三洋
f509b8c94a
Change the model configuration to trocr
2024-05-28 07:50:09 +00:00
三洋三洋
2ac159bfa2
Using paddleocr with onnxruntime
...
Deleted the code for test time.
2024-05-28 07:50:09 +00:00
三洋三洋
226c1e1f76
Added mixed recognition
...
change suryaocr to paddleocr
2024-05-28 07:50:08 +00:00
三洋三洋
a24ccd53ae
Added ONNX file for PaddleOCR model
2024-05-28 07:50:08 +00:00
三洋三洋
d3451d0ce7
Update .gitignore
2024-05-28 07:50:08 +00:00
三洋三洋
e2bf22dac8
Added code for PaddleOCR inference
2024-05-28 07:50:08 +00:00
三洋三洋
5c9cff2125
Eliminated dependency on paddleocr
...
Change to trocr
2024-05-28 07:50:08 +00:00
三洋三洋
cc602f5a82
update
2024-05-28 07:50:08 +00:00
OleehyO
19827f1837
bugfix: ocr_aug.py
...
Change "lhy_custom" in ink_swap_color to "random"
2024-05-28 07:49:55 +00:00
三洋三洋
0a51bde1c5
bugfix: missing filter_fn and inference/train transform
2024-05-12 07:49:04 +00:00